US8266510B1ActiveUtility
High-throughput pipelined and scalable architecture for a K-Best MIMO detector
Est. expiryNov 7, 2026(~0.3 yrs left)· nominal 20-yr term from priority
H04L 25/03203H04L 1/004H04L 25/03891
78
PatentIndex Score
7
Cited by
7
References
57
Claims
Abstract
A high throughput and scalable MIMO detector can use a K-Best detection algorithm to find K combinations of transmit symbols that are likely to be the symbols that were actually transmitted. The K-best MIMO detector can include a plurality of stages, where each stage may correspond to a transmit antenna, and each stage can find K best symbol combinations based on information from a previous stage. To find the new K best symbol combinations, at each stage, a plurality of metrics for potential combinations are computed and sorted by magnitude. The MIMO detector preferably uses a high throughput, merge sorting algorithm to sort the metrics.
Claims
exact text as granted — not AI-modified1. A method of reconstructing digital information in a multiple-input receiver from signals transmitted by a multiple-output transmitter, comprising:
obtaining a plurality of signals from a plurality of transmitted signals;
calculating a plurality of metrics for each transmitted signal based on the obtained signals;
sorting the metrics;
selecting a predetermined number of the sorted metrics; and
computing an estimate of the digital information based on the predetermined number of sorted metrics.
2. The method of claim 1 wherein the metrics are sorted using a merge sorting algorithm.
3. The method of claim 1 wherein the multiple-input receiver is a wireless receiver and the multiple-output transmitter is a wireless transmitter, and wherein the plurality of signals are derived from received signals received via receive antennas.
4. The method of claim 1 wherein each metric indicates a likelihood associated with a possible transmitted symbol.
5. The method of claim 4 wherein each metric is based on a distance to the associated symbol.
6. The method of claim 5 wherein each metric is computed according to
| Im{B i −R ii C j }|+|Re{B i −R ii C j }|,
where B i is an obtained signal, R ii is a component of an upper triangular matrix, and C j is a possible value of a transmitted signal.
7. The method of claim 1 wherein the predetermined number of sorted metrics are associated with combinations of symbols that are likely to correspond to actually transmitted symbols.
8. The method of claim 1 , further comprising triangularizing a channel response matrix associated with the obtained signals.
9. The method of claim 8 wherein triangularizing the channel response matrix comprises performing QR decomposition or Cholesky factorization on the channel response matrix.
10. The method of claim 1 wherein calculating the plurality of metrics comprises calculating the metrics in a plurality of stages, and wherein each stage is associated with an obtained signal.
11. The method of claim 10 wherein sorting the metrics comprises sorting the metrics within a same stage, the method further comprising selecting an intermediate set of metrics for each stage.
12. The method of claim 11 wherein calculating metrics comprises computing metrics based on an intermediate set of metrics from a previous stage.
13. The method of claim 10 wherein selecting the predetermined number comprises selecting metrics from a final stage.
14. The method of claim 10 wherein each obtained signal is based on a complex modulation scheme, and wherein each obtained signal is associated with:
a first stage for computing metrics based on real components of the complex modulation scheme; and
a second stage for computing metrics based on imaginary components of the complex modulation scheme.
15. The method of claim 14 , wherein the first stage computes metrics based on Re{y} and the second stage computes metrics based on Im{y}, where
[
Re
{
y
}
Im
{
y
}
]
=
[
Re
{
H
}
-
Im
{
H
}
Im
{
H
}
Re
{
H
}
]
[
Re
{
c
}
Im
{
c
}
]
+
[
Re
{
n
}
Im
{
n
}
]
,
y is a received signal vector, H is a channel response matrix, c is a transmitted signal vector, and n is a noise vector.
16. The method of claim 1 , further comprising:
determining a region associated with each obtained signal from a set of predefined regions; and
finding a subset of symbols associated with each region,
wherein calculating the plurality of metrics comprises calculating metrics for each symbol in the subsets of symbols.
17. The method of claim 16 wherein the subset of symbols comprises symbols that are likely to correspond to actually transmitted symbols.
18. The method of claim 1 wherein computing an estimate comprises computing soft information.
19. The method of claim 1 , further comprising decoding the digital information based on an error correcting code.
20. Apparatus for reconstructing digital information in a multiple-input receiver from signals transmitted by a multiple-output transmitter, comprising:
means for obtaining a plurality of signals from a plurality of transmitted signals;
means for calculating a plurality of metrics for each transmitted signal based on the obtained signals;
means for sorting the metrics;
means for selecting a predetermined number of the sorted metrics; and
means for computing an estimate of the digital information based on the predetermined number of sorted metrics.
21. The apparatus of claim 20 wherein the metrics are sorted using a merge sorting algorithm.
22. The apparatus of claim 20 wherein the multiple-input receiver is a wireless receiver and the multiple-output transmitter is a wireless transmitter, and wherein the plurality of signals are derived from received signals received via receive antennas.
23. The apparatus of claim 20 wherein each metric indicates a likelihood associated with a possible transmitted symbol.
24. The apparatus of claim 23 wherein each metric is based on a distance to the associated symbol.
25. The apparatus of claim 24 wherein each metric is computed according to
| Im{B i −R ii C j }|+|Re{B i −R ii C j }|,
where B i is an obtained signal, R ii is a component of an upper triangular matrix, and C j is a possible value of a transmitted signal.
26. The apparatus of claim 20 wherein the predetermined number of sorted metrics are associated with combinations of symbols that are likely to correspond to actually transmitted symbols.
27. The apparatus of claim 20 , further comprising means for triangularizing a channel response matrix associated with the obtained signals.
28. The apparatus of claim 27 wherein the means for triangularizing the channel response matrix comprises means for performing QR decomposition or Cholesky factorization on the channel response matrix.
29. The apparatus of claim 20 wherein the means for calculating the plurality of metrics comprises means for calculating the metrics in a plurality of stages, and wherein each stage is associated with an obtained signal.
30. The apparatus of claim 29 wherein the means for sorting the metrics comprises means for sorting the metrics within a same stage, the apparatus further comprising means for selecting an intermediate set of metrics for each stage.
31. The apparatus of claim 30 wherein the means for calculating metrics comprises means for computing metrics based on an intermediate set of metrics from a previous stage.
32. The apparatus of claim 29 wherein the means for selecting the predetermined number comprises means for selecting metrics from a final stage.
33. The apparatus of claim 29 wherein each obtained signal is based on a complex modulation scheme, and wherein each obtained signal is associated with:
a first stage for computing metrics based on real components of the complex modulation scheme; and
a second stage for computing metrics based on imaginary components of the complex modulation scheme.
34. The apparatus of claim 33 , wherein the first stage computes metrics based on Re{y} and the second stage computes metrics based on Im{y}, where
[
Re
{
y
}
Im
{
y
}
]
=
[
Re
{
H
}
-
Im
{
H
}
Im
{
H
}
Re
{
H
}
]
[
Re
{
c
}
Im
{
c
}
]
+
[
Re
{
n
}
Im
{
n
}
]
,
y is a received signal vector, H is a channel response matrix, c is a transmitted signal vector, and n is a noise vector.
35. The apparatus of claim 20 , further comprising:
means for determining a region associated with each obtained signal from a set of predefined regions; and
means for finding a subset of symbols associated with each region,
wherein the means for calculating the plurality of metrics means for comprises means for calculating metrics for each symbol in the subsets of symbols.
36. The apparatus of claim 35 wherein the subset of symbols comprises symbols that are most likely to correspond to actually transmitted symbols.
37. The apparatus of claim 20 wherein the means for computing an estimate comprises means for computing soft information.
38. The apparatus of claim 20 , further comprising means for decoding the digital information based on an error correcting code.
39. A receiver for reconstructing digital information from signals transmitted by a multiple-output transmitter, comprising:
a plurality of antennas configured to receive a plurality of signals from a plurality of transmit antennas; and
a detector, comprising:
a plurality of enumeration units configured to calculate a plurality of metrics for each transmitted signal based on processed signals derived from the received signals;
a plurality of sorting units configured to:
sort the plurality of metrics; and
select a predetermined number of sorted metrics; and
a computation unit configured to compute an estimate of the digital information based on the predetermined number of sorted metrics.
40. The receiver of claim 39 wherein the plurality of sorting units are further configured to sort the plurality of metrics based on a merge sorting algorithm.
41. The receiver of claim 39 wherein the receiver is a wireless receiver and the multiple-output transmitter is a wireless transmitter.
42. The receiver of claim 39 wherein each metric indicates a likelihood associated with a possible transmitted symbol.
43. The receiver of claim 42 wherein each metric is based on a distance to the associated symbol.
44. The receiver of claim 43 wherein the plurality of enumeration units are configured to compute metrics according to
| Im{B i −R ii C j }|+|Re{B i −R ii C j }|,
where B i is a obtained signal, R ii is a component of an upper triangular matrix, and C j is a possible value of a transmitted signal.
45. The receiver of claim 39 wherein the predetermined number of sorted metrics are associated with combinations of symbols that are most likely to correspond to actually transmitted symbols.
46. The receiver of claim 39 , further comprising a channel preprocessor configured to triangularize a channel response matrix associated with the received signals.
47. The receiver of claim 46 wherein the channel preprocessor is further configured to perform QR decomposition or Cholesky factorization on the channel response matrix.
48. The receiver of claim 39 wherein each enumeration unit is further configured to calculate metrics associated with one of the processed signals.
49. The receiver of claim 48 wherein each sorting unit is associated with an enumeration unit, wherein each sorting unit is further configured to select an intermediate set of metrics.
50. The receiver of claim 49 wherein an enumeration unit is further configured to calculate metrics based on an intermediate set of metrics from a previous sorting unit.
51. The receiver of claim 48 wherein the predetermined number of sorted metrics is selected from a final sorting unit.
52. The receiver of claim 48 wherein each processed signal is based on a complex modulation scheme, and wherein each processed signal is associated with:
a first one of the enumeration units configured to calculate metrics based on real components of the complex modulation scheme; and
a second one of the enumeration units configured to calculate metrics based on imaginary components of the complex modulation scheme.
53. The receiver of claim 52 , further comprising conversion circuitry configured to compute
[
Re
{
y
}
Im
{
y
}
]
=
[
Re
{
H
}
-
Im
{
H
}
Im
{
H
}
Re
{
H
}
]
[
Re
{
c
}
Im
{
c
}
]
+
[
Re
{
n
}
Im
{
n
}
]
,
where y is a received signal vector, H is a channel response matrix, c is a transmitted signal vector, and n is a noise vector.
54. The receiver of claim 39 wherein each enumeration unit further comprises:
a lookup table configured to:
determine a region associated with each processed signal from a set of predefined regions; and
find a subset of symbols associated with each region,
wherein the enumeration unit is configured to calculate metrics for each symbol in the subsets of symbols.
55. The receiver of claim 54 wherein the subset of symbols comprises symbols that are most likely to correspond to actually transmitted symbols.
56. The receiver of claim 39 wherein the computation unit is further configured to compute soft information.
57. The receiver of claim 39 , further comprising a decoder configured to decode the digital information based on an error correcting code.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.